1 files changed, 39 insertions, 20 deletions

diff --git a/dups b/dups
index 4f38482..58aae1c 100755
--- a/dups
+++ b/dups
@@ -25,9 +25,7 @@ my @preferences;
 # Map from size to list of filenames
 my %sizes;
 
-# Map from md5 to list of filenames
-my %md5s;
-
+# 'wanted' subroutine of File::Find::find. Add the file name to its size key in the %sizes hash
 sub register_file_size {
     return if -d "$_"; # Skip directories
 
@@ -37,11 +35,13 @@ sub register_file_size {
     push @{$sizes{$size}}, $filename;
 }
 
+# Return true iff '$file' is anywhere inside '$dir'
 sub indir($file, $dir) {
     my ($diffpath) = Cwd::abs_path($file) =~ /^$dir(.*)$/;
     return !!$diffpath;
 }
 
+# Take @files and filter it with the best results, discarding the ones from the disliked directories
 sub guess_best_choices(@files) {
     my @best_guesses = ();
 
@@ -69,7 +69,7 @@ sub keep($choice, @rest) {
     unlink foreach @delete;
 }
 
-print STDERR "Collecting file sizes...\n";
+# Parse CLI args
 while (my $arg = shift) {
     if ($arg eq "-y" || $arg eq "--yes") {
         $yes = 1;
@@ -78,29 +78,48 @@ while (my $arg = shift) {
     } elsif ($arg eq "-d" || $arg eq "--dislike") {
         push @preferences, {"type" => "dislike", "dir" => Cwd::abs_path(shift) . "/"};
     } else {
+        print STDERR "Collecting file sizes...\n";
         find(\&register_file_size, $arg);
     }
 }
 
-print STDERR "Computing md5s of files with same size...\n";
-my @progress = (0, 0, scalar(keys(%sizes)));
-foreach my $size (keys(%sizes)) {
-    $progress[0]++;
-    $progress[1] = 0;
-    my @same_size_files = @{$sizes{$size}};
-    next unless @same_size_files gt 1; # Discard unique sizes
-
-    foreach my $file (@same_size_files) {
-        $progress[1]++;
-        print STDERR "$progress[0].$progress[1] / $progress[2]\r";
-        STDERR->flush();
-
-        my ($md5) = capture("md5sum", "-z", $file) =~ m/^([^ ]+)/;
-        $md5s{$md5} ||= [];
-        push @{$md5s{$md5}}, $file;
+my $progressidx = 0;
+my $progressbar = '/-\|';
+sub progressbar($curr, $total) {
+    return sprintf "\r[%s] %s/%s", substr($progressbar, $progressidx++ % length($progressbar), 1), $curr, $total;
+}
+
+# Convert a hash of sizes to a hash of md5s
+sub sizes_to_md5s($sizes) {
+    my %md5s;
+    # progress: (curr_size, total_sizes)
+    my @progress = (1, scalar(keys(%$sizes)));
+    foreach my $size (keys(%$sizes)) {
+        my @same_size_files = @{$sizes->{$size}};
+
+        printf STDERR progressbar($progress[0], $progress[1]);
+        # Compute only md5 of non-unique sizes
+        if (@same_size_files gt 1) {
+            foreach my $file (@same_size_files) {
+                printf STDERR progressbar($progress[0], $progress[1]) unless $progress[1] == 0;
+                STDERR->flush();
+
+                my ($md5) = capture("md5sum", "-z", $file) =~ m/^([^ ]+)/;
+                $md5s{$md5} ||= [];
+                push @{$md5s{$md5}}, $file;
+            }
+        }
+
+        $progress[0]++;
     }
+
+    return \%md5s;
 }
 
+print STDERR "Computing md5s of files with same size...\n";
+# Map from md5 to list of filenames
+my %md5s = %{sizes_to_md5s(\%sizes)};
+
 print "\n";
 
 foreach my $md5 (keys(%md5s)) {